Hugging Face's logo

RMWeerasinghe
/
flan-t5-base-finetuned-QLoRA

Summarization
PEFT
Safetensors
Generated from Trainer
Model card Files Community
RMWeerasinghe commited on
Commit
1dc4195
·
verified ·
1 Parent(s): b2e1876

Training complete

Browse files
Files changed (2) hide show
  1. README.md +15 -16
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the cnn_dailymail dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: nan
25
- - Rouge1: 0.2276
26
- - Rouge2: 0.0926
27
- - Rougel: 0.1754
28
- - Rougelsum: 0.2072
29
 
30
  ## Model description
31
 
@@ -51,22 +51,21 @@ The following hyperparameters were used during training:
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - num_epochs: 10
54
- - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
60
- | 0.0 | 1.0 | 125 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
61
- | 0.0 | 2.0 | 250 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
62
- | 0.0 | 3.0 | 375 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
63
- | 0.0 | 4.0 | 500 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
64
- | 0.0 | 5.0 | 625 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
65
- | 0.0 | 6.0 | 750 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
66
- | 0.0 | 7.0 | 875 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
67
- | 0.0 | 8.0 | 1000 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
68
- | 0.0 | 9.0 | 1125 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
69
- | 0.0 | 10.0 | 1250 | nan | 0.2276 | 0.0926 | 0.1754 | 0.2072 |
70
 
71
 
72
  ### Framework versions
 
21
 
22
  This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the cnn_dailymail dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 1.5612
25
+ - Rouge1: 0.2459
26
+ - Rouge2: 0.1153
27
+ - Rougel: 0.1998
28
+ - Rougelsum: 0.2294
29
 
30
  ## Model description
31
 
 
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - num_epochs: 10
 
54
 
55
  ### Training results
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
58
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
59
+ | 15.0637 | 1.0 | 125 | 14.5380 | 0.2315 | 0.0949 | 0.1776 | 0.2084 |
60
+ | 11.8916 | 2.0 | 250 | 11.5849 | 0.2319 | 0.0957 | 0.1818 | 0.2077 |
61
+ | 7.8325 | 3.0 | 375 | 5.8491 | 0.2294 | 0.0978 | 0.1839 | 0.2098 |
62
+ | 3.6167 | 4.0 | 500 | 2.9519 | 0.2345 | 0.1096 | 0.1915 | 0.2195 |
63
+ | 3.0365 | 5.0 | 625 | 2.3695 | 0.2391 | 0.1132 | 0.1964 | 0.2228 |
64
+ | 2.7862 | 6.0 | 750 | 1.9609 | 0.2428 | 0.1163 | 0.1981 | 0.2258 |
65
+ | 2.5261 | 7.0 | 875 | 1.7701 | 0.2462 | 0.1137 | 0.1979 | 0.2292 |
66
+ | 2.3319 | 8.0 | 1000 | 1.6463 | 0.2459 | 0.1142 | 0.1995 | 0.2294 |
67
+ | 2.355 | 9.0 | 1125 | 1.5820 | 0.2442 | 0.114 | 0.1998 | 0.2279 |
68
+ | 2.323 | 10.0 | 1250 | 1.5612 | 0.2459 | 0.1153 | 0.1998 | 0.2294 |
69
 
70
 
71
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ffcf973bdc6d256456c3aa20fe15b286bbd69cc375a5af5e3f7ec3161cc8c505
3
  size 3558888
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a1b728dc2c3c11ae79cfa29bc653977b2191ad57dbd524e79860ecfc41eb30d
3
  size 3558888